Warfarin or dabigatran for treatment of atrial fibrillation.

BACKGROUND: New antithrombotic drugs for prevention and treatment of thromboembolic disorders in AF that are less demanding on local staff and facilities than warfarin should be welcomed if proved successful. OBJECTIVES: The comparative value and possible dangers of substituting the new drug dabigatran as a replacement remain to be established. Its safety and effectiveness must be reviewed and assessed by further study. METHODS: Clinical results of the European Action on Anticoagulation (EAA) computer-assisted dosage study and the Randomized Evaluation of Long-Term Anticoagulation Therapy (RE-LY) trial have been compared. RESULTS: Clinical events were lower in patients on warfarin in the EAA study compared to patients on both warfarin and dabigatran in the RE-LY study. CONCLUSION: Evaluations should recognize optimum requirements for safe and effective administration of both types of drug. In the warfarin arm improvements in effectiveness and safety recently introduced (i.e. the PT/INR line and variance growth analysis) should be included as they have been shown to be successful in improved prediction of bleeding and further thromboembolism. The incidence of bleeding with dabigatran, for which there is no antidote, will require evaluation.

The clinical evaluation of International Normalized Ratio variability and control in conventional oral anticoagulant administration by use of the variance growth rate.

INTRODUCTION: The time in target International Normalized Ratio (INR) range (TIR) is used to assess the control and intensity of oral anticoagulation, but it does not measure variation in the INR. OBJECTIVES: The value of assessing INR variability by use of the variance growth rate (VGR) as a predictor of events was investigated in patients treated with warfarin. METHODS: Three different methods of VGR determination (A, B1, and B2) together with the TIR were studied. Method A measures both INR variability and control, but methods B1 and B2 measure variability only. The VGR and TIR were determined over three time periods: overall follow-up to an event, and 6 months and 3 months before an event. RESULTS: Six hundred and sixty-one control patients were matched to 158 cases (bleeding, thromboembolism, or death). With all VGR methods, the risk of an event was greater in unstable patients at 6 months before an event than in stable patients. Method A demonstrated the greatest risk 3 months before an event in the unstable VGR group as compared with the stable group (odds ratio 3.3, 95% confidence interval 1.9-5.7, P < 0.005). The risk of an event was 1.9 times greater in patients with a low TIR (< 39%) than in those with a high TIR (> 80%) in the 3-month period (P = 0.02). Risk of bleeding was significantly greater in the 3-month period in patients with unstable VGR, with the greatest risk found with method B2 (P < 0.01). CONCLUSIONS: Patients with unstable anticoagulation have a significantly increased risk of 'clinical events' at 3 and 6 months before an event. The VGR can be incorporated into computer-dosage programs, and may offer additional safety when oral anticoagulation is monitored.

Coagulometer international sensitivity index (ISI) derivation, a rapid method using the prothrombin time/international normalized ratio (PT/INR) Line: a multicenter study.

BACKGROUND: The original WHO procedure for prothrombin time (PT) standardization has been almost entirely abandoned because of the universal use of PT coagulometers. These often give different international normalized ratio (INR) results from the manual method, between individual makes of instruments and with instruments from the same manufacture. METHOD: A simple procedure is required to derive local INR with coagulometers. The PT/INR Line method has recently been developed using five European Concerted Action on Anticoagulation (ECAA) certified plasmas to derive local INR. This procedure has been modified to derive a coagulometer PT/INR Line providing International Sensitivity Index (ISI) and mean normal PT (MNPT) for coagulometers and give local INR. Results have been compared with conventional ISI calibrations at the same laboratories. RESULTS: With human thromboplastins, mean ISI by local calibration was 0.93 (range: 0.77-1.16). With the PT/INR Line, mean coagulometer ISI was higher, for example 0.99 (0.84-1.23) but using the PT/INR Line derived MNPT there was no difference in local INR. Between-centre INR variation of a certified validation plasma was reduced with human and bovine reagents after correction with local ISI calibrations and the PT/INR Line. CONCLUSION: The PT/INR Line-ISI with its derived MNPT is shown to provide reliable local INR with the 13 different reagent/coagulometer combinations at the 28 centres in this international study.

The prothrombin time/international normalized ratio (PT/INR) Line: derivation of local INR with commercial thromboplastins and coagulometers--two independent studies.

BACKGROUND: The WHO scheme for prothrombin time (PT) standardization has been limited in application, because of its difficulties in implementation, particularly the need for mandatory manual PT testing and for local provision of thromboplastin international reference preparations (IRP). METHODS: The value of a new simpler procedure to derive international normalized ratio (INR), the PT/INR Line, based on only five European Concerted Action on Anticoagulation (ECAA) calibrant plasmas certified by experienced centres has been assessed in two independent exercises using a range of commercial thromboplastins and coagulometers. INRs were compared with manual certified values with thromboplastin IRP from expert centres and in the second study also with INRs from local ISI calibrations. RESULTS: In the first study with the PT/INR Line, 8.7% deviation from certified INRs was reduced to 1.1% with human reagents, and from 7.0% to 2.6% with rabbit reagents. In the second study, deviation was reduced from 11.2% to 0.4% with human reagents by both local ISI calibration and the PT/INR Line. With rabbit reagents, 10.4% deviation was reduced to 1.1% with both procedures; 4.9% deviation was reduced to 0.5% with bovine/combined reagents with local ISI calibrations and to 2.9% with the PT/INR Line. Mean INR dispersion was reduced with all thromboplastins and automated systems using the PT/INR Line. CONCLUSIONS: The procedure using the PT/INR Line provides reliable INR derivation without the need for WHO ISI calibration across the range of locally used commercial thromboplastins and automated PT systems included in two independent international studies.

The cost-effectiveness of computer-assisted anticoagulant dosage: results from the European Action on Anticoagulation (EAA) multicentre study.

BACKGROUND: Increased demand for oral anticoagulation has resulted in wider adoption of computer-assisted dosing in anticoagulant clinics. An economic evaluation has been performed to investigate the cost-effectiveness of computer-assisted dosing in comparison with manual dosing in patients on oral anticoagulant therapy. METHODS: A trial-based cost-effectiveness analysis was conducted as part of the EAA randomized study of computer-assisted dosage vs. manual dosing. The 4.5-year multinational trial was conducted in 32 centres with 13 219 anticoagulation patients randomized to manual or computer-assisted dosage. The main outcome measures were total health care costs, clinical event rates and cost-saving per clinical event prevented by computer dosing compared with manual dosing. RESULTS: Mean dosing costs per patient were lower (difference: euro47) for computer-assisted dosing, but with little difference in clinical event costs. Total overall costs were euro51 lower in the computer-assisted dosing arm. There were a larger number of clinical events in the manual dosing arm. The overall difference between trial arms was not significant (difference in clinical events, -0.003; 95% CI, -0.010-0.004) but there was a significant reduction in events with DVT/PE, suggesting computer-assisted dosage with the two study programs (dawn ac or parma 5) was at least as effective clinically as manual dosage. The cost-effectiveness analysis indicated that computer-assisted dosing is less costly than manual dosing. CONCLUSIONS: Results indicate that computer-assisted dosage with the two programs (dawn ac and parma 5) is cheaper than manual dosage and is at least as effective clinically, indicating that investment in this technology represents value for money.

An international multicenter randomized study of computer-assisted oral anticoagulant dosage vs. medical staff dosage.

BACKGROUND: Increased demand for oral anticoagulants is overwhelming facilities worldwide, resulting in increasing use of computer assistance. A multicenter clinical endpoint study has been performed to compare the safety and effectiveness of computer-assisted dosage with dosage by experienced medical staff at the same centers. METHODS: A randomized study of dosage of two commercial computer-assisted dosage programs (PARMA 5 and DAWN AC) vs. manual dosage at 32 centers with an established interest in oral anticoagulation in 13 countries. The aim was to recruit a minimum of 16,000 patient-years randomized to medical staff or computer-assisted dosage. In total, 13,219 patients participated, 6503 patients being randomized to medical staff and 6716 to computer-assisted dosage. The safety and effectiveness of computer-assisted dosage were compared with those of medical staff dosage. RESULTS: In total, 13,052 patients were recruited (18,617 patient-years). International Normalized Ratio (INR) tests numbered 193 890 with manual dosage and 193,424 with computer-assisted dosage. The number of clinical events with computer-assisted dosage was lower (P = 0.1), but in the 3209 patients with deep vein thrombosis/pulmonary embolism, they were reduced by 37 (24%, P = 0.001). Time in target INR range was significantly improved by computer assistance as compared with medical staff dosage at the majority of centers (P < 0.001). CONCLUSIONS: The safety and effectiveness of computer-assisted dosage has been demonstrated using two different marketed programs in comparison with experienced medical staff dosage at the centers with established interest in anticoagulation. Significant prevention of clinical events in patients with deep vein thrombosis/pulmonary embolism and the achievement of target INR in all clinical groups has been observed. The reliability and safety of other marketed computer-assisted dosage programs need to be established.